Dioxin, a ubiquitous contaminant of industrial combustion processes including medical waste incineration, has been implicated in the etiology of endometriosis in animals. We sought to determine whether dioxin exposure is associated with endometriosis in humans. We conducted a population-based historical cohort study 20 years after the 1976 factory explosion in Seveso, Italy, which resulted in the highest known population exposure to 2,3,7,8-tetrachlorodibenzo-(italic)p(/italic)-dioxin (TCDD). Participants were 601 female residents of the Seveso area who were (3/4) 30 years old in 1976 and had adequate stored sera. Endometriosis disease status was defined by pelvic surgery, current transvaginal ultrasound, pelvic examination, and interview (for history of infertility and pelvic pain). "Cases" were women who had surgically confirmed disease or an ultrasound consistent with endometriosis. "Nondiseased" women had surgery with no evidence of endometriosis or no signs or symptoms. Other women had uncertain status. To assess TCDD exposure, individual levels of TCDD were measured in stored sera collected soon after the accident. We identified 19 women with endometriosis and 277 nondiseased women. The relative risk ratios (RRRs) for women with serum TCDD levels of 20.1-100 ppt and >100 ppt were 1.2 [90% confidence interval (CI) = 0.3-4.5] and 2.1 (90% CI = 0.5-8.0), respectively, relative to women with TCDD levels (3/4) 20 ppt. Tests for trend using the above exposure categories and continuous log TCDD were nonsignificant. In conclusion, we report a doubled, nonsignificant risk for endometriosis among women with serum TCDD levels of 100 ppt or higher, but no clear dose response. Unavoidable disease misclassification in a population-based study may have led to an underestimate of the true risk of endometriosis.